Care management for serious mental illness: A systematic review and meta-analysis

Christopher Lim, MD

Cambridge Health Alliance
Care management for serious mental illness: A systematic review and meta-analysis

Scientific Abstract

Background: Persons with serious mental illness, encompassing schizophrenia spectrum disorders and bipolar disorder, face adverse psychiatric and medical outcomes and represent a large burden of healthcare cost. Care management, in which a provider offers assessment, care planning, and care coordination, is a common model of support, yet the evidence supporting its use among psychiatric populations is mixed. A systematic review and meta-analysis were undertaken to determine the impact of care management on clinical outcomes, acute care utilization, cost, and patient satisfaction among adults with serious mental illness.

Methods: A multi-database literature search was performed. Articles were included if they compared standard outpatient care with care management to standard outpatient care alone for adults with serious mental illness and reported on one or more predefined outcomes.

Randomized controlled trials and other study designs were permitted for inclusion. Meta-analysis included only randomized controlled trials.

Results: A total of 2,786 unique records were screened. For the qualitative synthesis, 34 papers representing 28 unique studies were included. Of these, 15 papers representing 12 unique studies were included for meta-analysis. In the meta-analysis, care management was associated with small, statistically significant favorable effect sizes for psychiatric symptoms, overall quality of life, and mental quality of life (Hedges’ g range 0.13–0.26). In addition, care management was associated with a small, statistically significant reduction in inpatient psychiatric hospital days (Hedges’ g=0.16).

Conclusions: Care management is associated with fewer psychiatric symptoms and greater quality of life for adults with serious mental illness. Further work is needed to determine which intervention components and contextual factors are associated with effectiveness.

SoundCloud Transcript

Hello, my name is Christopher Lim. I am a third-year psychiatry resident at Cambridge Health Alliance, and I am presenting a systematic review and meta-analysis on care management interventions for individuals with serious mental illness. Why did we conduct this study? Serious mental illness, which we define as encompassing schizophrenia spectrum disorders and bipolar disorder, are associated with early mortality, high burden of disability, and high health care costs. Care management is a common form of community support for chronic illnesses in which a provider performs assessment, care planning, and care coordination. Although intensive forms of care management such as Assertive Community Treatment have been shown to reduce hospitalizations and increase retention in care for individuals with serious mental illness, non-intensive forms of care management have had mixed results to date. By non-intensive care management, we are referring to care management that is non-team based, does not specify low staffing ratios, and does not require the direct provision of services. We conducted a systematic review and meta-analysis in accordance with the PRISMA guidelines to determine the impact of non-intensive care management (which we refer to in this study as simply “care management”) on individuals with serious mental illness across a range of outcomes. We performed a multi-database search. We included studies with a subject population of at least 50% schizophrenia spectrum disorder or bipolar disorder that compared standard outpatient care plus care management to standard outpatient care alone. We performed a random-effects meta-analysis on randomized controlled trials that were included and had adequate outcome data. In total, we screened 2,786 unique records, and yielded 28 unique studies in the systematic review and 12 unique studies for meta-analysis. A PRISMA diagram is in Figure 1. In terms of study characteristics, based on manuscript publication date, 15 studies were published in 2010 or later, 6 between 2000 and 2009, and 7 prior to 2000. Geographically, 10 studies were from the US, 12 from Europe, and 6 from other parts of the world. The 28 studies represented 12,783 subjects with an average age of 40 years, of which 44% were female. We also looked at characteristics of the care management interventions across the 28 studies. In all 28 studies, the intervention was at least in part in-person. In 17 studies, care management involved coordination of social services, in addition to coordination of clinical care. In 17 studies, the interventions involved direct provision of clinical services, such as psychoeducation, counseling on treatment, crisis intervention, or some other clinical skills or self-management training. Finally, in 19 studies, the interventions were exclusively delivered by a clinically trained individual such as a nurse or social worker. In the meta-analysis, we found care management was associated with an improvement in psychiatric symptoms, with a Hedges’ g of 0.15 and a p-value of 0.03. Care management was also associated with an improvement in global quality of life, with a Hedges’ g of 0.13 and a p-value of 0.002, and also an improvement in mental quality of life, with a Hedges’ g of 0.26 and a p-value of 0.04. Care management was associated with a reduction in inpatient psychiatric hospital days, with a Hedges’ g of 0.16 and a p-value of 0.02, but we did not find a statistically significant effect on total number of inpatient psychiatric hospitalizations. We found that the control, or standard outpatient care only, was associated with lower health care costs, with a Hedges’ g of 1.07 and a p-value of 0.02. Select forest plots are shown in Figure 2. In summary, based on the studies captured in this review, care management for serious mental illness has several common features, including being in-person, delivered by a clinician, involving social service coordination, and providing direct clinical service. Care management for serious mental illness is associated with small, favorable effects on psychiatric symptoms, global quality of life, and mental quality of life. Based on a sample of only 2 studies, care management for serious mental illness was associated with a large increase in health care costs. Care management was also associated with a small reduction in inpatient psychiatric hospital days, but not in total number of hospitalizations. Future directions for research include determining what patient, intervention, and contextual factors underlie the effectiveness of care management for serious mental illness, further assessment of the impact on health care utilization and cost, and cost-effectiveness analysis.

Live Zoom Session – April 21st

research Areas

Authors

Christopher T. Lim, M.D., Marissa P. Caan, M.D., Clara H. Kim, M.D., Clifton M. Chow, Ph.D., H. Stephen Leff, Ph.D., Miriam C. Tepper, M.D.